The left cerebral hemisphere in man contains anatomic structures specialized not only for language but also for higher-order motor programming. A major problem, however, in investigating the nature of this specialized movement control system has been the difficulty of obtaining objective measurement of movement in three-dimensional space. To this end, a series of investigations are proposed all involving three- dimensional computergraphic modeling and analysis of a neural disorder of movement (apraxia) that results from failure of this left hemisphere system. Analysis and simulations of impairments in movement timing, spatial relations, and joint control will provide unique insight into the nature of the left cerebral hemisphere's representations of learned, skilled movements. The investigation proceeds along four lines of inquiry: Fractional of Neural Representations of Learned Movement. Alternate models of the basis of apraxia are tested, focussing on the hypothesis that apraxia results from destruction of spatiotemporal representations of learned movement stored in the left hemisphere. The fractionation of these representations due to specific brain lesions are then investigated. Formation of New Spatiotemporal Representations. The brain's formation of new spatiotemporal representations after destruction of existing ones is investigated both through the analysis of the pattern of recovery of function in apraxic subjects and through study of the fractionation of the gesture acquisition process. Loss of Motor Equivalence in Apraxia. The consequences of the destruction of spatiotemporal representations of learned movement for the flexible control of behavior are explored by investigating the possible loss of motor equivalence in apraxic subjects. We hypothesize that apraxic subjects, but not subjects with right hemisphere lesions, will have lost the flexible control of motor behavior that motor equivalence entails. Computer Simulations of Apraxic Errors. The spatiotemporal representations of learned movement are computergraphically modeled, and the movement errors that apraxic subjects make simulated by 'lesioning' component processes in the model. In this manner, the control processes that may be disrupted in apraxia can be illuminated. By gaining insight into the nature of the representations of learned, skilled movement in the brain, new ways to compensate for their disruption by stroke or other forms of brain damage can be developed. These studies, taken as a whole, should provide unique insights into the basis of apraxia, the fractionation of spatiotemporal representations of learned, skilled movement following specific brain lesions, and into the specialized role of the left hemisphere for higher order motor control.